Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations (readOnlyHint=true, destructiveHint=false, etc.) already establish this is a safe read operation, and the description is fully consistent with them—no contradiction exists. The description adds a modest behavioral detail by clarifying the traversal direction (upward) and the terminal concept ('ultimate beneficial owner(s)'), which goes beyond the tin name. However, it does not disclose behavior around cycles, max_depth enforcement, or how incomplete ownership stakes are handled, so it adds limited incremental context for a graph walk tool.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Is the description appropriately sized, front-loaded, and free of redundancy?

One sentence of roughly 15 words, front-loaded with the verb, and every word earns its place by communicating direction (upward), starting point (root entity), and endpoint (beneficial owner(s)). Zero fluff or filler.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a relatively simple 2-parameter flat tool with no nested objects and no output schema, the description covers the essential mechanism adequately, and the rubric notes the description needn't explain return values without an output schema. However, given the graph-traversal nature of the tool and the existence of similarly-shaped siblings (graph_path, graph_neighbourhood), the description misses key operational details like default depth, cycle handling, and the threshold for 'beneficial owner'—details that would materially affect agent behavior.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 0%, so the description was expected to carry the full explanatory burden for the parameters, but it only loosely echoes the word 'root' from the schema without explaining what a valid root value is or what the 'max_depth' parameter does, its units, or default. A tool that includes both an unbounded 'root' and an optional numeric 'max_depth' leaves real ambiguity (e.g., direction of depth, default behavior when absent) that the description fails to resolve. The description adds almost no parameter-level meaning beyond the schema's bare type declarations.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Does the description clearly state what the tool does and how it differs from similar tools?

The description uses a specific verb ('walk') plus a clear resource ('ownership... upward from a root entity to the ultimate beneficial owner(s)') with a directional scope (upward) and a defined terminus (beneficial owners). This is a clear statement of purpose that incidentally differentiates the tool from the sibling `graph_path` (generic pathfinding) and `get_entity` (single-entity lookup), though it doesn't name alternatives. The language is specific enough to avoid confusion with most siblings.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

No explicit when-to-use guidance, exclusions, or named alternatives are provided despite a rich sibling context (graph_path, graph_neighbourhood, find_similar) where such guidance would be valuable. The 'walk upward' phrasing weakly implies a use case (finding ultimate corporate ownership), but there is no statement about when NOT to use this tool or which sibling would be a better fit, leaving the agent to guess between this and graph_path.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.